ModelsMedium importanceContext
ElevenLabs launches Scribe v3 and Scribe v3 Live, transcription that uses the conversation's context to avoid cutting callers off mid-sentence
ElevenLabs launched two speech-to-text models, Scribe v3, for recorded audio, and Scribe v3 Live, for real-time use, designed for difficult audio such as people talking over each other or background noise. According to the company, Scribe v3 makes about 25% fewer word errors than its predecessor across 102 languages, and speaker separation now works live. The Live version analyzes the context of the conversation, in addition to the sound, to know when the person has finished speaking, instead of cutting in at a pause. In the Artificial Analysis test, Scribe v3 is among the most accurate on the market, behind only an Alibaba preview, but it is slower and more expensive than the Microsoft and xAI models. The company also expanded its partnership with NVIDIA and announced an alliance with McKinsey.
Published in edition no. 26 of Noticiário do Joãozinho ·
Translated by Joãozinho AI from the original Brazilian Portuguese edition. Facts, numbers and sources are the same as in the original; quotes, video titles and source names stay in their original language. Read the original in Portuguese
Original source
https://the-decoder.com/elevenlabs-scribe-v3-live-uses-conversational-context-to-stop-voice-agents-from-cutting-off-callers-mid-sentence/